Chronology of alluvial terrace sediment accumulation and incision … · 30 Fluvial sediments originating from mountain belts like the Andes yield important archives of 31 past environmental
Post on 27-Sep-2020
0 Views
Preview:
Transcript
Chronology of alluvial terrace sediment accumulation and 1
incision in the Pativilca Valley, western Peruvian Andes 2
3
Camille Litty1*, Fritz Schlunegger1, Naki Akçar1, Romain Delunel1, Marcus Christl2, 4
Christof Vockenhuber2 5
1 Institute of Geological Sciences, University of Bern, Baltzerstrasse 1+3, CH- 3012 Bern. 6
2Laboratory of Ion Beam Physics, ETH Zurich, Zurich, Switzerland 7
* Current address: Univ. Grenoble Alpes, IUGA, ISTerre, 38000 Grenoble, France 8
ABSTRACT 9
The incision and aggradation of the Pativilca alluvial fan delta system in the western Peruvian 10
Andes through Quaternary time can be traced in detail using well-exposed fill terraces studied by a 11
combination of cosmogenic nuclide dating, terrace mapping and paleo-erosion rate calculations. Two 12
alluvial terraces have been dated through depth-profile exposure dating using in-situ 10Be. The dating 13
results return an age for the abandonment of the terrace at 200 r 90 ka in Pativilca and 1.2 Ma r 0.3 14
Ma in Barranca. These new ages complete the database of previously dated terrace fills in the valley. 15
Together with the results of the terrace mapping and the absolute ages of the terraces, we show that 16
the valley fills are made up of at least four terraces; two terraces near the city of Pativilca and two 17
terraces in the city of Barranca. While previous studies have shown two periods of sediment 18
aggradation, one period around 100 ka (Barranca) and another period around 30 ka (Pativilca), our 19
new results show two additional periods of sediment aggradation and subsequent incision that have 20
not been reported before. Finally, paleo-erosion rates at the time of the deposition of the terrace 21
material were calculated and compared to the available modern estimates. The paleo-erosion rates 22
vary from 140 ± 12 m/Ma to 390 ± 40 m/Ma. The period of sediment accumulation prior to the 23
abandonment of the terrace at 200 ka corresponds to a wet phase and a pulse of erosion. In contrast, 24
the period of sediment accumulation prior to the abandonment of the terrace at 1.2 Ma does not 25
source: https://doi.org/10.7892/boris.116603 | downloaded: 29.1.2021
correspond to a pulse of erosion and could rather correspond to a change of the base level possibly 26
induced by a sea-level rise. 27
Keywords: 10Be depth-profile dating; alluvial terraces; Pativilca Valley; Western Peruvian Andes 28
1. Introduction 29
Fluvial sediments originating from mountain belts like the Andes yield important archives of 30
past environmental or tectonic changes. The sediments can record changes in precipitation rates and 31
climate (Litty et al., 2016; d’Arcy et al., 2017). They can also record the response to earthquake-32
induced landslides (McPhillips et al., 2014). The reconstruction of the timing of alluvial sediment 33
deposition thus bears important information when the scope lies in the detection of specific climate or 34
tectonic events as driving forces of landscape evolution. In this context, depth-profile dating based on 35
in-situ produced 10Be measured in quartz has been proven a reliable method to establish a chronology 36
of sediment deposition (e.g., Bookhagen et al., 2006; Hidy et al., 2010). In particular, this 37
methodology yields an age when sediment aggradation stopped and when a period of sediment 38
accumulation was superseded by a phase of erosion and incision into the previously deposited 39
material. 10Be is the most commonly measured in situ–produced cosmogenic nuclide (Granger et al., 40
2013). Its dominance in geological applications stems from several factors, including the abundance 41
of the target mineral, quartz, a standardized chemistry procedure (Kohl and Nishiizumi, 1992), a 42
relatively simple production depth profile, and routinely good precision by accelerator mass 43
spectrometry (AMS) (Granger, 2006). Additionally, isochron burial dating using 10Be and 26Al is 44
becoming increasingly important in studies related to river terraces (e.g., Darling et al., 2012; Erlanger 45
et al., 2012; Akçar et al., 2017). Isochron-burial dating yields in an age when the investigated material 46
accumulated. It is thus a variation of traditional burial dating methods. Ages, or alternatively the 47
burial times of sediment, are determined using the difference between the cosmogenic 26Al/10Be 48
surface production ratio at the time of burial and the 26Al/10Be ratio measured in buried sediments 49
(Granger, 2006). Sediments of alluvial terrace deposits with flat tops are ideal for surface exposure 50
dating and isochron burial dating: they are persistent, easily identifiable as surfaces that were formed 51
at a specific time and that have been isolated from the fluvial system since deposition. Because they 52
are typically coarse-grained, well drained, and nearly flat, they can be remarkably well preserved and 53
unaffected by erosion, especially in arid environments like in the western side of the Peruvian Andes 54
(Litty et al., 2017a; Reber et al., 2017). 55
Alluvial terrace sequences are common features along the coastal margin between Peru and 56
northern Chile. They are located particularly in lower valley reaches near to the Pacific coast (Steffen 57
et al., 2009, 2010; Trauerstein et al., 2014; Litty et al., 2017a). Climate change has been considered to 58
have controlled pulses of erosion on the western Andean margin through the increase in mean surface 59
runoff resulting either in sediment accumulation along stream segments close to the Pacific coast 60
(Bekaddour et al., 2014; Norton et al., 2016), or in surface erosion in upstream segments of major 61
rivers (Veit et al., 2016). These climate-driven changes have been interpreted as being the main 62
driving force controlling the sediment accumulation and the formation of cut-and-fill terraces on the 63
western Andean margin (Norton et al., 2016). In the Pativilca Valley, situated on the western margin 64
of the Peruvian Andes at about 10°S (Fig. 1A), a terrace sequence has been previously dated through 65
infrared stimulated luminescence (IRSL) techniques (Trauerstein et al., 2014). The results have 66
disclosed the occurrence of at least two periods of sediment aggradation, one period spanning from 10 67
ka to 90 ka with an age of the samples cluster around 30 ka and another period spanning the time 68
interval between 80 ka and 130 ka with an age estimate of the samples cluster around 110 ka 69
(Trauerstein et al., 2014). While generally wetter climate results in fluvial incision, the results from 70
Trauerstein et al. (2014) suggested that here wetter climate conditions do correlate with periods of 71
fluvial aggradation. 72
The aim of this study is to date additional terrace deposits using in-situ terrestrial cosmogenic 73
nuclides to complete the chronological framework and to infer the history of sediment aggradation 74
and incision in this alluvial fan delta system. In addition concentrations of in-situ 10Be recorded by 75
detrital quartz minerals in the terrace deposits will be used to infer the paleo-erosion rates recorded at 76
the time when sediment accumulation occurred. These rates will be compared to the modern ones 77
(Reber et al., 2017) to quantify the erosion in the upstream drainage basin during the phases of 78
aggradation within the downstream valley. The final aim is to understand the factors controlling 79
fluvial aggradation and incision in fan delta environments in the western Andes. 80
81
2. Regional settings 82
The Pativilca Valley is located in central Peru, about 200 km to the northwest of Lima. The 83
Rio Pativilca, which is trunk stream of the region, debouches into the Pacific at 10.7°S and 77.8°W 84
(Fig. 1A). The drainage basin has an area of about 4400 km2, and the longest flow path measures 85
approximately 200 km. The upper section of the stream is characterized by a bedrock channel with a 86
steep gradient (knickzone), whereas in the lower segment the narrow valley floor is covered by 87
alluvial deposits that are thickening and widening towards the coast, giving way to an alluvial fan 88
delta. The sedimentological architecture of the deposits is characterized by amalgamated stacks of 20 89
to 50 m-thick units of poorly sorted, clast-supported conglomerates with a coarse-grained sandy 90
matrix (Fig. 1B). The clasts are subrounded and sometimes imbricated, but the sedimentary fabrics are 91
predominantly massive (Fig. 1B). The alluvial conglomerates are part of an alluvial fan delta system 92
characterized by a suite of individual fill terraces with different altitudes of the tread (Fig. 1B). 93
The precipitation pattern of South America is strongly influenced by the low level Andean jet 94
and the position of the Inter Tropical Convergence Zone (ITCZ), which experiences seasonal shifts in 95
response to insolation differences between austral summer and winter. The Andean jet transfers 96
humidity from the Pacific Ocean and the Amazon basin to the eastern margin of the Andes, and also 97
to the Altiplano and the western Andean margin (Garreaud, 2009). The Andean mountain range thus 98
acts as a major topographic barrier to the atmospheric circulation. As a result of this circulation 99
pattern, the Peruvian western margin shows an E-W contrasting precipitation pattern with high annual 100
precipitation rates up to 800 mm on the Altiplano and ~0 mm along the coast. From north to south, the 101
annual rainfall rates on the Altiplano decrease from 1000 mm near the Equator to <200 mm in 102
northern Chile. Every 2-10 yr, near the Equator, the Pacific coast is subjected to stronger precipitation 103
than the mean precipitation rates, resulting in high flood magnitude variability related to the El Nino 104
Southern Oscillation (ENSO) weather phenomenon (DeVries, 1987). Today, this phenomenon is 105
limited to the coastal area of northern Peru, but during the past, southern Peru might also have been 106
affected by such events (Lagos et al., 2008). 107
On orbital time scales, the position of the ITCZ has shifted in response to larger insolation 108
and heat contrasts between the Northern and Southern Hemispheres, which has been related to the 109
effects of shifts in the Earth’s precession (Strecker et al., 2007). The results are stronger upper air 110
easterlies and more precipitation on the Altiplano (Garreaud et al., 2003). Variations in precipitation 111
rates and patterns led to remarkable lake level variations on the Altiplano as recorded by lake level 112
highstands on the plateau (Ouki, Minchin and Tauca pluvial periods, e.g., Fritz et al., 2004). These 113
climate changes have also controlled pulses of erosion and deposition on the western Andean margin 114
(Bekaddour et al., 2014; Veit et al., 2016). Related variations in erosional fluxes have been interpreted 115
as being the main factor controlling the formation of cut-and-fill terrace systems along the western 116
margin of the Peruvian Andes (Norton et al., 2016). 117
118
3. Methods 119
3.1. Cosmogenic nuclides 120
Over the past 25 yr, cosmogenic nuclides have become an essential tool in Quaternary 121
geochronology (e.g., Gosse and Phillips, 2001; Granger, 2006). Cosmogenic nuclides are produced 122
through spallation reactions and muon capture in minerals of rocks and sediment at or near the Earth’s 123
surface (Gosse and Philips, 2001). Cosmogenic 10Be and 26Al can be applied to determine a post-124
depositional age of a geological layer using their accumulation (depth-profile dating). Alternatively, 125
they can also be used to determine the timing of sediment accumulation through their radioactive 126
decay (burial dating) history (e.g., Anderson et al., 1996; Repka et al., 1997; Granger and Smith, 127
2000; Granger and Muzikar, 2001; Wolkowinsky and Granger, 2004; Balco and Rovey, 2008; Akçar 128
et al., 2017). 129
Depth-profile dating is based on the exponential decrease of cosmogenic nuclides with depth 130
(Gosse and Philips, 2001). On the other hand, the burial dating technique uses the difference in half-131
lives of 10Be (1.387 Ma; Korschinek et al., 2010; Chmeleff et al., 2010) and 26Al (0.705 Ma; Norris et 132
al., 1983) and thus the 26Al versus 10Be ratio to determine the burial time, when the pre-burial and 133
post-burial concentrations are known or estimated (e.g., Granger and Muzikar, 2001; Akçar et al., 134
2017). We followed the Erlanger et al. (2012) isochron approach where one of the advantages is the 135
assumption that post-burial production is identical across a single stratigraphic horizon. 136
The collected samples (see section 3.3 for description of sample sites and sampling strategy) 137
were processed in the Surface Exposure Laboratory of the Institute of Geological Sciences at the 138
University of Bern following the lab protocol described in Akçar et al. (2012). The 10Be/9Be and 139
26Al/27Al AMS measurements were then performed at the Swiss Federal Institute of Technology 140
tandem facility in Zurich (Christl et al., 2013). The long-term weighted average 10Be/9Be ratio of (2.41 141
± 0.53) × 10−15 was used for full process blank correction. Table 1 presents the samples information 142
and cosmogenic nuclide results. 143
Depth-profile ages were modelled with MATLAB® using Monte Carlo simulations 144
developed by Hidy et al. (2010). Depth-profile patterns were simulated based on exposure age, 145
erosion rate and inheritance. Table 2 shows the input parameters for the Barranca and Pativilca depth-146
profile simulations. We applied no correction factor for topographic shielding. We justify this 147
approach because there is no significant topography around the sampling sites that could block a 148
portion of incoming cosmic radiations (Dunne et al., 1999; Gosse and Phillips, 2001), as the sampling 149
sites are located on the widest and flattest part of the valley close to the coast. We did not consider 150
snow cover to have a major impact on the results as the mean basin elevation of the sampled 151
catchment is largely situated below the snow line. The 10Be half-life with a value of 1.387 ± 0.012 Ma 152
was utilized (Chmeleff et al., 2010; Korschinek et al., 2010). The local production rate was scaled to 153
the Lal (1991) and Stone (2000) scheme using a production rate caused by spallation (SLHL: at sea-154
level, high latitude) of 4.01 ± 0.12 atoms gSiO2-1 (CRONUS calculator update from v. 2.2 to v. 2.3 155
published by Balco in August 2016 after Balco et al., 2008; Borchers et al., 2016). Thus a site-specific 156
spallogenic production rate of 2.5 ± 0.5 atoms g-1 a-1 was obtained for Barranca and for Pativilca. We 157
applied a bulk density ranging between 1.6 and 2.1 g cm-3 for the sediment samples in Barranca and 158
Pativilca. Finally, to model a depth-profile age we simulated 100,000 profiles and used a χ2 cut-off 159
value of ≤ 20 for Barranca and ≤ 3 for Pativilca (Table 2). 160
161
3.2. Paleo erosion rates 162
Paleo basin-averaged erosion rates can be calculated using the cosmogenic nuclide 163
concentrations of past sediment samples following Granger et al. (1996) and von Blanckenburg 164
(2005). To calculate the basin averaged paleo-erosion rate, we used the 10Be cosmogenic nuclide 165
concentrations of the sand embedded in the terrace deposits after corrections have been made for 166
shielding, post-depositional nuclide production at sample depth z, and atom loss due to radioactive 167
decay during time t (both considered in Eq. (1); Balco et al., 2008). These equations can be used 168
assuming: (i) The material was well mixed in the upstream basin and finally embedded in the terrace 169
fill. This appears to be the case in the western Peruvian valleys where the fluvial processes have 170
dominated the transport of sediment (Litty et al., 2017b), thus providing well-mixed material. (ii) The 171
paleo-erosion is representative for the entire catchment. Indeed, the sediments of the Pleistocene 172
terrace fills in western Peru record an origin from both the upper flat part of the catchments and the 173
lower steep reaches (Litty et al., 2017a). (iii) The residence of the material on the hillslopes and the 174
channels is much shorter than the erosional timescale. This is the case in the western Peruvian valleys 175
where regolith was considered to have been rapidly stripped from hillslopes, which most likely 176
resulted in the supply of large volumes of sediment to the trunk streams during the periods of 177
sediment aggradation (Norton et al., 2016). (iv) The individual terraces have not experienced multiple 178
phases of erosion and re-deposition, so that major internal unconformities are not present (von 179
Blanckenburg, 2005). This appears to be the case in the Pativilca Valley as no unconformities in the 180
individual terrace fills have been observed in the field. 181
182
3.3. Sampling sites 183
Two previously undated alluvial terrace fills were sampled for depth-profile exposure dating. 184
These terraces are located along the lowermost reach of the Pativilca River and in the city of Barranca 185
(Fig. 1; Table 1). At each sampling site, six samples were collected along a vertical profile from 0.9 to 186
4.7 m beneath the tread of the terrace in Pativilca, and from 0.4 to 3.2 m beneath the tread of the 187
terrace in Barranca (Fig. 1A). Two to three kilograms of medium grained sand embedded between the 188
pebbles were taken for each sample. Additionally, the lowermost samples of the two depth profiles 189
(PAT-DP6 and BAR-DP6) were used to infer a paleo-erosion rate at the time when the sediments of 190
the two newly dated terraces were deposited. Two other samples (PAT-PE and BAR-PE2) were 191
collected in two other terrace fills previously dated (Trauerstein et al., 2014) for the calculation of 192
paleo-erosion rates (one in Pativilca and one in Barranca; Table 4). Additionally, quartz bearing clasts 193
were sampled for isochron burial dating (Fig. 1B; Table 1). For each isochron burial site, the samples 194
were collected from the same sedimentologic unit and from a single stratigraphic horizon following 195
Erlanger et al. (2012). Three horizons were sampled in Barranca and two horizons have been sampled 196
in Pativilca (Fig. 1B; Table 1). Depth-profile dating and isochron burial dating techniques have be 197
chosen as sand lenses that are required for IRSL sampling are not present in every terrace fill. 198
199
4. Results 200
4.1. Cosmogenic nuclides: isochron burial dating 201
The measured 26Al concentrations are plotted versus 10Be concentrations including 2σ 202
uncertainties (Fig. 2). The cosmogenic nuclide results are shown in Table 1. As the Al/Be ratios are 203
higher than the surface ratio, it is not possible to calculate an isochron burial age from these samples 204
(for details, see Erlanger et al., 2012). The surface ratio of 26Al/10Be is not constant since it depends 205
on the time of exposure and erosion. On a banana-plot, the ratios decrease from 8.4 to ~3, and a 206
regression through these yields a surface ratio around 6.8. Therefore, in most of the isochron burial 207
applications this ratio has been used as the surface ratio. Recently, Akçar et al. (2017) showed that 208
this ratio varied between 7 and 12 in deeply eroding landscapes, particularly in glacial environments. 209
However, these mechanisms fail to explain the 26Al/10Be ratios > 12 obtained in this study as glacial 210
processes were most likely not the most important erosional mechanisms. Therefore, we tentatively 211
attribute these ratios to the analytical problems related to the measurements of the total Al or to the 212
quartz purification process. Given that no age can be determined from these samples; isochron burial 213
dating is therefore not further discussed in this paper. 214
215
4.2. Cosmogenic nuclides: depth-profile dating 216
4.2.1. Barranca 217
AMS-measured 10Be/9Be (with uncertainties) as well as calculated 10Be concentrations for 218
each sample are shown in Table 1. The concentrations of the six sediment samples vary from ~12 x 219
105 atoms g-1 for the uppermost sample to ~1 x 105 atoms g-1 for the lowermost sample (Table 1). In 220
Fig. 3, the 10Be concentrations together with 1σ uncertainties are plotted against depth. They display 221
an exponential decrease with depth. The simulated best fit curve through the six data points is 222
illustrated in Fig. 4, whereas the possible solution space with a χ2 cut-off value of ≤ 20 is shown in 223
Fig. 5. 224
The simulation yields a best-fit solution to the measured nuclide concentrations for a modal 225
depth-profile age of 1.2 r 0.3 Ma, and a modal top erosion rate of 0.07 r 0.02 cm ka-1 (Table 3A). The 226
modal values of the age and erosion rate are similar to the mean and median values of the simulation, 227
thus the errors of the modal values are based on the minimum and maximum values generated by the 228
simulation. Note that the Monte Carlo simulation code requires a constraint on the net erosion on the 229
top of the section as a modal input parameter to calculate an age (Hidy et al., 2010). This parameter is 230
iteratively adjusted within a range of values. 231
232
4.2.2. Pativilca 233
The concentrations of the six sediment samples vary from ~86 x 104 atoms g-1 for the 234
uppermost sample to ~44 x 104 atoms g-1 for the lowermost sample (Table 1). In Fig. 6, the 10Be 235
concentrations together with 1σ uncertainties are plotted against depth. The best fit through the six 236
data points is illustrated in Fig. 7, whereas the possible solution space with a χ2 cut-off value of ≤ 3 is 237
shown in Fig. 8. 238
The simulation yields a best-fit solution to the measured nuclide concentrations for a modal 239
depth-profile age of 200 r 90 ka, a modal top erosion of 0.48+ 0.41- 0.13 cm ka-1 and an inheritance 240
of 35,100+ 8700-8000 atoms g-1 (Table 3B). The modal values are similar to the mean and median 241
values of the simulation, thus the errors of the modal values are based on the minimum and maximum 242
values generated by the simulation. 243
The results of the depth-profile dating return a surface exposure age of ~1.2 Ma in Barranca 244
and of ~200 ka in Pativilca. These results show minimum ages when the accumulation of material has 245
terminated and when dissection of the previously deposited material started, yielding in the formation 246
of a terrace level. These two periods when sediment aggradation was superseded by dissection, have 247
not been dated before. These new results together with the ones from Trauestein et al. (2014) suggest 248
the occurrence of at least four terraces referred to as T1 to T4 from older to younger (Figs. 9 and 249
10A), corresponding to at least four different periods of sediment accumulation. Terrace deposits were 250
correlated on the basis of landscape position, tread altitude and absolute dating (Figs. 9 and 10A). 251
252
4.3. Paleo-erosion rates 253
The in-situ 10Be analytical data together with the inferred paleo-erosion rates recorded by the 254
alluvial terrace sediments are presented in Table 4. The paleo-erosion rate values are 143 r 12 m Ma-1 255
at the time of the accumulation of the terrace deposits T1 (~1.2 Ma ago), 302 r 28 m Ma-1 at the time 256
when terrace material T2 was deposited, 392 r 40 m Ma-1 at the time of the deposition of the terrace 257
sediments T3, and finally 297 r 29 m Ma-1 at the time terrace T4 was constructed (Fig. 10B). In 258
addition, Reber et al. (2017) reported a modern catchment-averaged denudation rate of 260 r 23 m 259
Ma-1. 260
5. Discussion 261
5.1. Chronology of sediment accumulation and incision 262
The fluvial aggradation and subsequent incision in the Pativilca Valley has occurred in 263
multiple episodes through the Quaternary (Figs. 9 and 10). Figure 11 shows the position of the active 264
river and the position of the sediment accumulation during the periods of aggradation. Our dating 265
results imply that the sediments of terrace T1 in Barranca have been deposited prior to 1.2 Ma. The 266
aggradation then ceased and the tread formation began ~1.2 Ma ago (terrace T1; Fig. 11A). During 267
the period of the terrace fill, the erosion rate was two times lower than the modern rate (Fig. 12). 268
Following this, for approximately 1 Ma, either a period of no sedimentation occurred in the valley or 269
no sediments have been preserved. The river then moved its course towards Pativilca. The sediments 270
of terrace T2 in Pativilca have been deposited prior to 200 ka. The accumulation of sediment then 271
stopped and exposed the terrace tread at around 200 ka (terrace T2; Fig. 11B). During the period of 272
sediment accumulation, the erosion rate was up to ~300 m.Ma-1 (Fig. 12). The river bed again 273
changed its course towards Barranca, and a phase of accumulation occurred around 100 ka (deposition 274
of the sediment of T3; Fig. 11C; Trauerstein et al., 2014). In this period, the erosion rate was at its 275
highest (~400 m Ma-1; Fig. 12). Finally, the lobe of the Pativilca fan delta moved back towards the 276
city of Pativilca close to its current course, and a phase of aggradation occurred from 10 to 45 ka ago 277
(deposition of the sediment of T4; Fig. 11D; Trauerstein et al., 2014). During this period, the 278
catchment-wide denudation rate dropped back to ~300 m Ma-1 (Fig. 12). This phase was followed by 279
a period of incision exposing the tread and riser of terrace level T4. Today the erosion rate is slightly 280
lower than during the past at ~200 ka (Fig. 12) and the river appears to be incising. 281
282
5.2. Implications for climate variability as controls on cyclic deposition and erosion 283
A stratigraphic record of river terrace sediments is formed and preserved as a stream changes 284
its activity between incision, lateral planation, and aggradation (Pederson et al., 2006). These fill 285
terraces in the Pativilca Valley represent a relatively complete archive of both incision and deposition. 286
They can be used to understand the response to climate or to other driving forces that have an impact 287
on the balance between sediment transport and deposition (Pederson et al., 2006). These terrace fills 288
have been formed in the alluvial fan delta of the Pativilca River and they might also record the change 289
in the position of the different lobes of the delta through shifts in transport and sediment capacity. 290
Alternatively, a phase of accumulation requires the availability of sediments on the hillslopes to be 291
eroded, transported and deposited (Hancock and Anderson, 2002). This implies that the river 292
experiences an increase in the ratio between sediment supply and the stream’s capacity to control the 293
deposition the supplied material (Tucker and Slingerland, 1997). The youngest period of sediment 294
accumulation ranging from 10 to 45 ka (terrace T4) could correspond to the wet intervals recorded by 295
the Minchin (47.8–36 ka ago) and Tauca (26–14.9 ka ago) paleolakes (Fritz et al., 2004). The period 296
of sediment accumulation ranging from 80 to 130 ka (terrace T3) could correspond to the wet period 297
characterized by the Ouki paleolakes (120-98 ka ago; Fritz et al., 2004). These two periods of 298
sediment accumulation previously dated by Trauerstein et al. (2014) are thus correlated with phases of 299
enhanced precipitation with higher water discharge in the river. These wet conditions could have been 300
induced by summer insolation forcing of the South American summer monsoon at precessional time-301
scales (Baker et al., 2001a,b). Indeed, the precession together with the obliquity has been considered 302
to control the seasonal cycles of insolation (Milankovitch, 1941). The wettest phases, and hence the 303
highest lake levels (Bills et al., 1994; Sylvestre et al., 1999; Placzek et al., 2006), were additionally 304
forced by warm North Atlantic sea surface temperatures (Baker et al., 2001a). These climate changes 305
were also used to explain the pulses of upland erosion and deposition in the stream valleys on the 306
western Andean margin (Bekaddour et al., 2014), which agree with our data of relative fast paleo-307
erosion rates recorded by these two terraces. Indeed, the 10-45 ka denudation rate was >10% higher 308
than the modern one and the 80-130 ka denudation rate was even >30% higher than the modern rates 309
(Fig. 12). Fluvial aggradation is here correlated with wetter climates and an increased sediment supply 310
from the uplands. However, we also note that wetter climates can result in fluvial incision and terrace 311
formation because of greater stream discharge (Veit et al., 2016), provided that the hillslopes have 312
been depleted of material (Norton et al., 2016). In our case, the start of the incision phases could then 313
correspond to the end of the pluvial period and the time of decrease of the supply of sediment to the 314
river. Alternatively, it is also possible that erosional recycling of the terrace material started within the 315
pluvial periods, when the preceding phase of rapid hillslope erosion resulted in the depletion of the 316
sediment reservoirs, yielding high ratios between water and sediment fluxes in the trunk stream. The 317
Altiplano lake sediment cores do not record climatic variations older than 130 ka (Placzek et al., 318
2006). In this context, we cannot correlate the two older periods of sediment accumulation (prior to 319
~200 ka and prior to ~1.2 Ma) to any lake level variations. However, the high paleo-erosion rate 320
calculated for the newly dated fills of the terrace T2 (~15% higher than the modern one) appears also 321
to correspond to a pulse of upland erosion, which could point towards a period of wet conditions. 322
Support for this interpretation is provided by the periodicity of about 100 ka for this orbital-induced 323
summer insolation forcing (Milankovitch, 1941; Lisiecki, 2010; Abe-Ouchi et al., 2013). If this 324
interpretation is valid, then the ages of 100 ka (T3) and 200 ka (T2) would then correspond to this 100 325
ka periodicity, suggesting that the period prior to ~200 ka might also have corresponded to a wet 326
phase on the Altiplano. The oldest dated period of sediment accumulation (terrace T1) does not record 327
a distinct pulse of erosion as the calculated paleo erosion rate was twice as low as the modern one. 328
The production of sediments on the hillslopes through weathering and erosion can occur through an 329
increase in precipitation rates (e.g., Bookhagen et al., 2005; Norton et al., 2016), for which there is no 330
evidence from the records reported here indicating that sediment accumulation of the terrace T1 has 331
not been induced by a phase of enhanced precipitation. Alternatively, this T1 phase of accumulation 332
could have occurred in response to a rise in sea level. Indeed, a rise in sea level would cause a back 333
filling and a super-elevation of the channel, which then would cause the delta lobe to switch positions. 334
Supporting evidence for this interpretation has been provided by Pillans et al. (1998), who proposed 335
that the ~1.2 Ma-old period was relatively warm and corresponded to a rising sea level. Nevertheless, 336
we note that further research in the region is required to sustain this interpretation. 337
338
6. Conclusions 339
The results of the depth-profile dating together with the previously published IRSL ages 340
(Trauerstein et al., 2014) disclose at least four terraces located in the fan delta of the Pativilca River. 341
The sediments of terrace T1 accumulated until ~1.2 Ma ago and terraces T2, T3 and T4 were 342
deposited prior to ~200 ka, ~100 ka ago and ~30 ka ago respectively. Additionally, paleo-erosion 343
rates at the time of the deposition of the terrace fills were calculated and compared to the modern 344
rates. The modern erosion rate is ~260 mm/ka, while the paleo-erosion rates vary from ~143 mm/ka to 345
~391 m Ma-1. 346
The oldest period of accumulation does not correspond to a distinct pulse of erosion and could 347
rather correspond to a period when the sea level was rising. The three younger phases of sediment 348
accumulation most likely correspond to wet phases and pulses of erosion in the uplands. These wet 349
conditions were likely to have been induced by summer insolation forcing of the South American 350
summer monsoon at precessional time scales (Baker et al., 2001a, 2001b). Generally, wetter climate 351
results in fluvial incision caused by greater stream discharge. However, wetter climate is here 352
correlated with fluvial aggradation due to the inferred increased sediment supply from the uplands. 353
The abandonment of the terrace treads would then correspond to the end of the pluvial period and thus 354
a decrease of the sediment supplied to the river. Additionally, this long period of preservation of the 355
alluvial sediments on a coastal area implies a constant base level after the deposition of the terrace T1 356
despite the occurrence of an active subduction zone. 357
358
Acknowledgements 359 Support in the laboratory by Julia Krbanjevic is greatly appreciated. The support by Prof. Silvia Rosas 360 and her team (PUCP, Lima) to solve logistic problems at the customs are greatly acknolwedged. This 361 research has been supported by funds of the Swiss National Science Foundation awarded to 362 Schlunegger (project number 155892). 363 364 References 365 366 Abe-Ouchi, A., Saito, F., Kawamura, K., Raymo, M. E., Okuno, J. I., Takahashi, K., Blatter, H., 2013. 367 Insolation-driven 100,000-year glacial cycles and hysteresis of ice-sheet volume. Nature 500(7461), 368 190-193. 369 370 Akçar, N., Deline, P., Ivy-Ochs, S., Alfimov, V., Hajdas, I., Kubik, P.W., et al., 2012. The 1717 AD 371 rock avalanche deposits in the upper Ferret Valley (Italy): a dating approach with cosmogenic 10Be. 372 Journal of Quaternary Science 27(4), 337–440. 373 374 Akçar, N., Ivy-Ochs, S., Alfimov, V., Schlunegger, F., Claude, A., Reber, R., Christl, M., 375 Vockenhuber, C., Dehnert, A., Rahn M., Schlüchter, C. 2017. Isochron-burial dating of glaciofluvial 376 deposits: First results from the Swiss Alps. Earth Surface Processes Landforms. In Press. 377 378 Anderson, R.S., Repka, J.L., Dick, G.S., 1996. Explicit treatment of inheritance in dating depositional 379 surfaces using in situ 10Be and 26Al. Geology 24(1), 47-51. 380 381 Baker P.A., Seltzer G.O., Fritz, S.C., Dunbar, R.B., Grove, M.J., Tapia, P.M., Cross, S.L., Rowe, 382 H.D., Broda, J.P. 2001a. The history of South American tropical precipitation for the past 25,000 383 years. Science 291,640–643. 384 385 Baker, P.A., Rigsby, C.A., Seltzer, G.O., Fritz, S.C., Lowenstein, T., Bacher, N., Veliz, C. 2001b. 386 Tropical climate changes at millennial and orbital timescales on the Bolivian Altiplano. Nature 409, 387 698–701. 388 389 Balco, G., Rovey, C.W., 2008. An isochron method for cosmogenicnuclide dating of buried soils and 390 sediments. American Journal of Sciences 308, 1083–1114. 391 392 Balco, G., Stone, J.O., Lifton, N.A., Dunai, T.J., 2008. A complete and easily accessible means of 393 calculating surface exposure ages or erosion rates from Be-10 and Al-26 measurements. Quarternary 394 Geochronology 3, 174–195. 395 396 Bekaddour, T., Schlunegger, F., Vogel, H., Delunel, R., Norton, K. P., Akçar, N., Kubik, P., 2014. 397 Paleo erosion rates and climate shifts recorded by Quaternary cut-and-fill sequences in the Pisco 398 Valley, central Peru. Earth and planetary science letters 390, 103-115. 399
400 Bills, B.G., De Silva, S.L., Currey, D.R., Emenger, R.S., Lillquist, K.D., Donnellan, A., Worden, B., 401 1994. Hydro‐isostatic deflection and tectonic tilting in the central Andes: Initial results of a GPS 402 survey of Lake Minchin shorelines. Geophysical Research Letters 21(4), 293-296. 403 404 Bookhagen, B., Thiede, R. C., Strecker, M. R., 2005. Abnormal monsoon years and their control on 405 erosion and sediment flux in the high, arid northwest Himalaya. Earth and Planetary Science Letters, 406 231(1), 131-146. 407 408 Bookhagen, B., Fleitmann, D., Nishiizumi, K., Strecker, M. R., Thiede, R. C., 2006. Holocene 409 monsoonal dynamics and fluvial terrace formation in the northwest Himalaya, India. Geology 34(7), 410 601-604. 411 412 Borchers, B., Marrero, S., Balco, G., Caffee, M., Goehring, B., Lifton, N., Nishiizumi, K., Phillips, F., 413 Schaefer, J., Stone, J., 2016. Geological calibration of spallation production rates in the CRONUS-414 Earth project. Quaternary Geochronology 31, 188–198. 415 416 Braucher, R., Merchel, S., Borgomano, J., Bourles, D.L., 2011. Production of cosmogenic 417 radionuclides at great depth: A multi element approach. Earth and Planetary Science Letters 309(1), 418 1–9. 419 420 Chmeleff, J., von Blanckenburg, F., Kossert, K., Jakob, D., 2010. Determination of the Be-10 half-life 421 by multicollector ICP-MS and liquid scintillation counting. Nuclear Instruments and Methods in 422 Physics, Research Section B: Beam Interactions with Materials and Atoms 268. 423 424 Christl, M., Vockenhuber, C., Kubik, P.W., Wacker, L., Lachner, J., Alfimov, V., Synal, H.A., 2013. 425 The ETH Zurich AMS facilities: Performance parameters and reference materials. Nuclear 426 Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and 427 Atoms 294, 29-38. 428 429 D'arcy, M., Whittaker, A.C., Roda‐Boluda, D.C., 2017. Measuring alluvial fan sensitivity to past 430 climate changes using a self‐similarity approach to grain‐size fining, Death Valley, California. 431 Sedimentology 64(2), 388-424. 432 433 Darling, A.L., Karlstrom, K.E., Granger, D.E., Aslan, A., Kirby, E., Ouimet, W. B., Cole, R. D., 2012. 434 New incision rates along the Colorado River system based on cosmogenic burial dating of terraces: 435 Implications for regional controls on Quaternary incision. Geosphere 8(5), 1020-1041. 436 437 DeVries, T.J., 1987, A review of geological evidence for ancient El Nino activity in Peru. J. Geophys. 438 Res 92(14), 471–14 439 440 Dunne, J., Elmore, D., Muzikar, P., 1999. Scaling factors for the rates of production of cosmogenic 441 nuclides for geometric shielding and attenuation at depth on sloped surfaces. Geomorphology 27, 3–442 11. 443 444 Erlanger, E.D., Granger, D.E., Gibbon, R. J., 2012. Rock uplift rates in South Africa from isochron 445 burial dating of fluvial and marine terraces. Geology 40(11), 1019-1022. 446 447
Fritz, S.C., Baker, P.A., Lowenstein, T.K., Seltzer, G.O., Rigsby, C.A., Dwyer, G.S., and Luo, S., 448 2004, Hydrologic variation during the last 170,000 years in the southern hemisphere tropics of South 449 America: Quarternary Research 61, 95–104, doi: 10 .1016 /j .yqres .2003 .08 .007. 450 451 Garreaud, R. D., 2009. The Andes climate and weather. Advances in Geosciences 22, 3. 452 453 Garreaud, R., Vuille, M., Clement, A. C., 2003. The climate of the Altiplano: Observed current 454 conditions and mechanisms of past changes. Paleogeography, Paleoclimatology, Paleoecology 194, 455 5–22. 456 457 Gosse, J.C., Phillips, F.M., 2001. Terrestrial in situ cosmogenic nuclides: theory and application. 458 Quaternary Science Reviews 20(14), 1475-1560. 459 460 Granger DE. 2006. A review of burial dating methods using 26Al and 10Be. Geological 461 Society of America Special Publication 415, 1-16. 462 463 Granger, D.E., Smith, A.L., 2000. Dating buried sediments using radioactive decay and muogenic 464 production of 26Al and 10Be. Nuclear Instruments and Methods in Physics Research, B172, 822-826. 465 466 Granger, D.E., Kirchner, J.W., Finkel, R.C., 1996. Spatially averaged long-term erosion rates 467 measured from in situ-produced cosmogenic nuclides in alluvial sediment. Journal of Geology 104, 468 249-257. 469 470 Granger, D.E., Muzikar, P.F., 2001. Dating sediment burial with in situproduced cosmogenic 471 nuclides: theory, techniques, and limitations. Earth and Planetary Science Letters 188, 269–281. 472 473 Granger, D.E., Lifton, N.A., Willenbring, J.K., 2013. A cosmic trip: 25 years of cosmogenic nuclides 474 in geology. Geological Society of America Bulletin 125(9-10), 1379-1402. 475 476 Hancock, G.S., Anderson, R.S., 2002. Numerical modeling of fluvial strath-terrace formation in 477 response to oscillating climate. Geological Society of America Bulletin 114(9), 1131-1142. 478 479 Hidy, A.J., Gosse, J.C., Pederson, J.L., Mattern, J.P., Finkel, R.C., 2010. A geologically constrained 480 Monte Carlo approach to modeling exposure ages from profiles of cosmogenic nuclides: An example 481 from Lees Ferry, Arizona. Geochemistry Geophysics, Geosystems 11. 482 483 Kohl, C.P., Nishiizumi, K., 1992. Chemical isolation of quartz for measurement of in-situ-produced 484 cosmogenic nuclides. Geochimica et Cosmochimica Acta 56(9), 3583-3587. 485 486 Korschinek, G., Bergmaier, A., Faestermann, T., Gerstmann, U.C., Knie, K., Rugel, G., 2010. A new 487 value for the half-life of Be-10 by heavy-ion elastic recoil detection and liquid scintillation counting. 488 Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials 489 and Atoms 268, 187-191. 490 491 Lagos, P., Silva, Y., Nickl, E., Mosquera, K., 2008. El Niño? related precipitation variability in Perú. 492 Advances in Geosciences 14, 231-237. 493 494
Lal, D., 1991. Cosmic-ray labeling of erosion surfaces—In situ nuclide production-rates and erosion 495 models. Earth and Planetary Science Letters 104, 424–439. 496 497 Lisiecki, L.E., 2010. Links between eccentricity forcing and the 100,000-year glacial cycle. Nature 498 Geosci. 3, 349–352. 499 500 Litty C., Duller R., Schlunegger F., 2016. Paleohydraulic reconstruction of a 40 kyr-old terrace 501 sequence implies that water discharge was larger than today. Earth Surface Processes and Landforms 502 41, 884-898. DOI: 10.1002/esp.3872. 503 504 Litty, C., Lanari, P., Burn, M., Schlunegger, F., 2017a. Climate-controlled shifts in sediment 505 provenance inferred from detrital zircon ages, western Peruvian Andes. Geology 45(1), 59-62. 506 507 Litty, C., Schlunegger, F., Viveen, W., 2017b. Possible threshold controls on sediment grain 508 properties of Peruvian coastal river basins. Earth Surface Dynamics 5, 1-13. 509 https://doi.org/10.5194/esurf-5-1-2017 510 511 McPhillips, D., Bierman, P.R., Rood, D. H., 2014. Millennial-scale record of landslides in the 512 Andesconsistent with earthquake trigger. Nature Geoscience 7(12), 925-930. 513 514 NASA Land Processes Distributed Active Archive Center (LP DAAC). ASTER GDEM. USGS/Earth 515 Resources Observation and Science (EROS) Center, Sioux Falls, South Dakota, 2001. 516 517 Norris, T.L., Gancarz, A.J., Rokop, D.J., Thomas, K.W., 1983. Half‐life of 26Al. Journal of 518 Geophysical Research, Solid Earth 88(S01). 519 520 Norton, K.P., Schlunegger, F., Litty. C., 2016. On the potential for regolith control of fluvial terrace 521 formation in semi-arid escarpments. Earth Surf. Dynam 4, 147-157, https://doi.org/10.5194/esurf-4-522 147-2016. 523 524 Milankovitch, M. Kanon der Erdbestrahlung und seine Anwendung auf das Eiszeitproblem (R. 525 Serbian Acad., 1941). 526 527 Pederson, J.L., Anders, M.D., Rittenhour, T.M., Sharp, W.D., Gosse, J. C., Karlstrom, K.E., 2006. 528 Using fill terraces to understand incision rates and evolution of the Colorado River in eastern Grand 529 Canyon, Arizona. Journal of Geophysical Research: Earth Surface 111(F2). 530 531 Pillans, B., Chappell, J., Naish, T.R., 1998. A review of the Milankovitch climatic beat: template for 532 Plio–Pleistocene sea-level changes and sequence stratigraphy. Sedimentary Geology 122(1), 5-21. 533 534 Placzek, C., Quade, J., Patchett, P.J., 2006. Geochronology and stratigraphy of late Pleistocene lake 535 cycles on the southern Bolivian Altiplano: implications for causes of tropical climate change. 536 Geological Society of America Bulletin 118, 515–532. 537 538 Reber, R., Delunel, R., Schlunegger, F., Litty, C., Madella, A., Akçar, N., Christl, M., 2017. 539 Environmental controls on 10Be‐based catchment‐averaged denudation rates along the western 540 margin of the Peruvian Andes. Terra Nova. 541 542
Repka, J.L., Anderson, R.S., Finkel, R.C., 1997. Cosmogenic dating of fluvial terraces, Fremont 543 River, Utah. Earth and Planetary Science Letters 152(1), 59-73. 544 545 Steffen, D., Schlunegger, F., Preusser, F., 2009. Drainage basin response to climate change in the 546 Pisco Valley, Peru. Geology 37, 491–494. 547 548 Steffen, D., Schlunegger, F., Preusser, F., 2010, Late Pleistocene fans and terraces in the Majes 549 Valley, southern Peru, and their relation to climatic variations. International Journal of Earth Sciences 550 99(8), 1975-1989. 551 552 Stone, J.O., 2000. Air pressure and cosmogenic isotope production. Journal of Geophysical Research: 553 Solid Earth 105(B10), 23753-23759. 554 555 Strecker, M.R., Alonso, R.N., Bookhagen, B., Carrapa, B., Hilley, G.E., Sobel, E.R., Trauth, M.H., 556 2007, Tectonics and climate of the southern central Andes: Annual Review of Earth and Planetary 557 Sciences 35, 747–787, doi: 10 .1146 /annurev .earth .35 .031306 .140158. 558 559 Sylvestre, F., Servant, M., Servant-Vildary, S., Causse, C., Fournier, M., Ybert, J. P., 1999. Lake-560 level chronology on the Southern Bolivian Altiplano (18–23 S) during late-glacial time and the early 561 Holocene. Quaternary Research 51(1), 54-66. 562 563 Trauerstein M, Lowick SE, Preusser F, Schlunegger F. 2014. Small aliquot and single grain IRSL and 564 post-IR IRSL dating of fluvial and alluvial sediments from the Pativilca Valley, Peru. Quaternary 565 Geochronology 112, 163–174. 566 567 Tucker GE, Slingerland R. 1997. Drainage basin response to climate change. Water Resources 568 Researc. 33, 2031–2047. 569 570 Veit, H., May, J.H., Madella, A., Delunel, R., Schlunegger, F., Szidat, S., Capriles, J. M., 2016. 571 Palaeo‐ geoecological significance of Pleistocene trees in the Lluta Valley, Atacama Desert. Journal 572 of Quaternary Science 31(3), 203-213. 573 574 Von Blanckenburg F. 2005. The control mechanisms of erosion and weathering at basin scale from 575 cosmogenic nuclides in river sediment. Earth and Planetary Science Letters 237(3), 462–479. 576 577 Wolkowinsky, A.J., Granger, D.E., 2004. Early Pleistocene incision of the San Juan River, Utah, 578 dated with 26Al and 10Be. Geology 32(9), 749-752. 579 580 581 Tables and Figures captions 582 583 Table 1: Sample and cosmogenic nuclide data. 584 585 Table 2: Input parameters for the Monte Carlo simulator in Matlab® (Hidy et al., 2010). 586 587 Table 3: Results of the Monte Carlo simulations with Matlab® for (A) Barranca and (B) Pativilca. 588 Total number of simulated profiles is 100,000. The bold numbers represent the modelled values and 589 are therefore the ones that are used in this paper. 590
591 Table 4: Information relevant for interpreting 10Be concentrations. Modern and paleo catchment-592 averaged denudation were calculated using the SRTM DEM with a 90 m resolution. A 10Be half-life 593 of 1.39 +/- 0.01 Ma was used (Chmeleff et al., 2010; Korschinek et al., 2010) and a SLHL 10Be 594 production rate of 4.01 at g-1 a-1. A density of 2.65 g cm-3 was employed. 595 596 Fig. 1: (A) Maps of the study area showing the location of Pativilca and Barranca on the western side 597 of the Peruvian Andes. (B) Field photographs showing the alluvial terraces in Barranca and Pativilca. 598 Samples PAT-DP-1 to 6 (Pativilca), BAR-DP 1 to 6 (Barranca) were collected for depth-profile 599 dating. The white lines represent the bracket level where quartz bearing clasts were sampled for 600 isochron burial dating purposes. The concentrations obtained for the samples BAR-DP6 and PAT-601 DP6 were used for the calculation of the paleo-basin wide denudation rates 602 603 Fig. 2: Measured 26Al concentrations plotted vs. 10Be concentrations of the isochron-burial dating 604 samples in Barranca (BAR-IS1, BAR-IS2 and BAR-IS3) and in Pativilca (PAT-IS1 and PAT-IS2). 605 The sampling sites are shown on Fig. 1B. The errors represent 2σ uncertainties. The dash lines 606 illustrate the surface production rate ratio of 6.75 (Balco et al., 2008). 607 608 Fig. 3: Measured 10Be concentrations including the 1σ uncertainties of the Barranca depth-profile 609 samples plotted against depth. 610 611 Fig. 4: Modal output of the Monte Carlo simulations showing frequency distributions and χ2 values 612 for exposure age, erosion rate and inheritance. 613 614 Fig. 5: Output of the Monte Carlo depth-profile age simulation. (A) Illustration of the best fit through 615 the samples for the lowest χ2 value. (B) Possible solution space with a χ2 cut-off values of < 20. 616 617 Fig. 6: Measured 10Be concentrations including the 1σ uncertainties of the Pativilca depth-profile 618 samples plotted against depth. 619 620 Fig. 7: Modal output of the Monte Carlo simulations showing frequency distributions and χ2 values 621 for exposure age, erosion rate and inheritance. 622 623 Fig. 8: Output of the Monte Carlo depth-profile age simulation. (A) Illustration of the best fit through 624 the samples for the lowest χ2 value. (B) Possible solution space with a χ2 cut-off values of < 3. 625 626 Fig. 9: Map of the alluvial terraces in Pativilca and Barranca showing the age of the different terraces. 627 628 Fig. 10: (A) Summary of the IRSL and depth-profile ages in Pativilca and Barranca. The black dots 629 represent the IRSL samples from Trauerstein et al. (2014) and the white dot represents the depth 630 profile (this study). (B) Summary of the paleo-catchment wide denudation rates.The transect A-B can 631 be seen in Fig. 9. 632 633 Fig. 11: Maps showing the inferred channel belt position during the sediment accumulation phases in 634 the Pativilca Valley. (A) prior to ~1.2 Ma ago. (B) prior to ~200 ka ago. (C) ~100 ka ago. (D) ~30 ka 635 ago. 636 637 Fig. 12: Erosion rates versus time. 638
BAR-DP 1 BAR-DP 2BAR-DP 3
BAR-DP 4BAR-DP 5
BAR-DP 6
Depth profile sample 2 m
53 m.a.s.l.
OSL sample
Pativilca 1
1
2
2
2 m
95 masl75 masl
PAT-DP5PAT-DP4PAT-DP3
PAT-DP6
PAT-DP2PAT-DP1
OSL sample
Isochron sample level
BAR-IS1 (50m) BAR-IS2(33m)
BAR-IS3(25m)
PAT-IS1(90m)
PAT-IS2
2
Pisco
Sampling site
±10.5°S
10°S
77°W77.5°WBarrancaPativilca0 m
6730 m a.s.l.
A
B
Barranca
Pativilca
0
500000
1500000
2500000
0 100000 200000 300000 400000
PativilcaSurface ratio 6.75
0
20000
60000
100000
140000
0 4000 8000 12000
PAT-IS2-33.86 mg
PAT-IS1-13.79 mg
PAT-IS1-22.9 mg
PAT-IS2-72.18 mg
PAT-IS2-14.33 mg
PAT-IS1-93.79 mg
PAT-IS1-74.1 mg
PAT-IS2-44.32 mg
PAT-IS2-23.93 mg
PAT-IS2-63.77 mg
Surface ratio 6.75
10Be (atom/g)
10Be (atom/g)
26A
l (at
om/g
)
26A
l (at
om/g
)
0
400000
800000
1200000
1600000
2000000
0 20000 60000 100000 140000
Barranca
Surface ratio 6.75
BAR-IS2-43.94 mg
BAR-IS3-2 (remeasure)4.09 mg
BAR-IS2-6 2.96 mg
BAR-IS3-24.09 mg
BAR-IS3-4 3.42 mg
BAR-IS2-1 3.2 mg
BAR-IS1-23.55 mg
BAR-IS1-63.03 mg
BAR-IS1-5 2.35 mg
BAR-IS1-72.29 mg
BAR-IS1-33.62 mg
BAR-IS1-44.04 mg
10Be (atom/g)
26A
l (at
om/g
)
Dep
th (c
m)
10Be concentration (atom/g)
0
50
100
150
200
250
300
350
Barranca Depth Profile (Peru)
50 m a.s.l.
200,000 600,000 1,000,000
Concentration (atoms g-1)10^6 Concentration (atoms g-1)10^60 1 2 3 10 2 3
Dep
th (c
m)
Dep
th (c
m)150
50
100
200
250
300
350
00
150
50
100
200
250
300
350
Concentration vs. depth Concentration vs. depthA. B.
A B
Dep
th (c
m)
0
100
200
300
400
500
600,000200,000 400,000 800,000
95 m a.s.l.
10Be concentration (atom/g)
Dep
th (c
m)
150
50
100
200
250
300
350
0
400
450
500
Concentration (atoms g-1)10^4
4 6 8 10
Concentration (atoms g-1)10^4
4 6 8 10
Dep
th (c
m)
150
50
100
200
250
300
350
0
400
450
500
Concentration vs. depth Concentration vs. depthB.A.
BA
Pativilca
Barranca
T1
T2Depth profile + Paleo-erosion rateModern mean catchment erosion rate
Pativilca
Barranca
T2
T4
Depth profile (This study)
5 km
OSL sample (Trauerstein et al., 2014)
T4
N
110-140 ka
25-45 ka 10-40 ka
80-130 ka?
T2 20-90 ka
A
B
200+/-86ka
1.2 +/- 0.3 MaT1
T3
C
D
Figure 9
T4T2 T4
exposure age:200 +/- 86 ka
burial age:25-45 ka
burial age:10-40 ka
95 m a.s.l.75 m a.s.l.
120 m a.s.l.A B
Depth profile (This study)
OSL sample (Trauerstein et al., 2014)
70 m
?
exposure age:1.2 +/- 0.3 Ma
52 m a.s.l.T131 m a.s.l.
?
burial age:80-130 ka
?
T3
Not to scale
Pativilca Barranca
modern channel
?
C D
T4T2 T4
paleo erosion rate : 302 +/- 28mm/ka
95 m a.s.l.75 m a.s.l.
90 m a.s.l.
A B
70 m
?
52 m a.s.l.T131 m a.s.l.
??
T3
Not to scale
Pativilca Barranca
modern erosionrate: 260 +/- 23 mm/ka*
?
paleo erosion rate : 297 +/- 29mm/ka
paleo erosion rate : 143 +/- 12mm/ka
paleo erosion rate : 392 +/- 40 mm/ka*Reber et al (2017)
C D
A. Ages
B. Erosion rates
C) c. 110 ka ago
Pativilca Pativilca
Barranca
D) c. 30 ka ago
A) prior to 1.2 Ma
Pativilca Pativilca
Barranca
B) prior to 200 ka
N
5 km
Barranca
Barranca
1200200 400 600 800 1000
Time (ka)
Ero
sion
rate
(mm
/ka)
100
200
300
400
0 1400
Table 1. Sample information and cosmogenic nuclide data of the samples
Site Technics Latitude (°) Longitude (°)Altitude of the
top of the terrace (m)
Sample name Sample depth(cm)
Sample type Quartz dissolved(g)
9Be spike(mg) 10Be/9BeRelative
uncertainty(%)
10Be concentration (10^4 atoms/g)
Al(mg) 26Al/27AlRelative
uncertainty(%)
26Al concentration (10^4 atoms/g)
26Al/10Be Error on the ratio
Pativilca Depth Profile dating
-10.704 -77.775 96 PAT-DP1 90-100 Sand 15.0271 0.1731 1,14E-13 8.72 8.59 n.d n.d. n.d. n.d. n.d. n.d.
PAT-DP2 130-140 49.8738 0.1702 2,68E-13 5.23 6.06PAT-DP3 170-180 49.8325 0.1745 2,98E-13 5.18 6.91PAT-DP4 210-220 50.0394 0.175 2,71E-13 5.38 6.26PAT-DP5 270-280 49.9998 0.1725 1,96E-13 7.08 4.46PAT-DP6 460-470 49.8804 0.1743 2,43E-13 4.64 5.62
Isochron burial dating
-10.704 -77.775 96 PAT-IS1-1 460-470 Quartz bearing clasts
38.207 0.1739 5,79E-13 18.4 17.5 3.79 3,46E-13 17,5 76.7 4.38 1.11
PAT-IS1-2 38.7454 0.1737 6,45E-14 22.9 1.86 2.90 2,30E-13 18,6 38.5 20.7 6.24PAT-IS1-7 36.6074 0.1747 1,40E-14 9.96 0.368 4.10 2,03E-14 33,7 5.07 13.8 4.92PAT-IS1-9 37.689 0.1746 1,68E-14 9.81 0.443 3.79 1,76E-14 37,8 3.96 8.92 3.52
Isochron burial dating
-10.708 -77.772 75 PAT-IS2-1 400-410 Quartz bearing clasts
39.5056 0.1737 1,32E-14 16.0 0.315 4.33 1,66E-14 39,3 4.06 12.8 5.65
PAT-IS2-3 40.9181 0.1739 1,41E-12 2.73 3.98 3.86 1,25E-12 2,78 263 6.62 0.26PAT-IS2-7 36.3518 0.175 9,91E-15 15.6 0.241 2.18 1,89E-14 32,8 2.53 10.5 4.08PAT-IS2-2 32.5973 0.2004 2,39E-14 11.5 0.884 3.94 3,25E-14 10,6 8.77 9.93 1.65PAT-IS2-4 39.2653 0.1972 2,05E-14 10.6 0.608 4.32 3,47E-14 11,6 8.52 14.0 2.34PAT-IS2-6 39.8664 0.1998 4,47E-14 8.79 1.41 3.78 5,44E-14 8,79 11.5 8.12 1.04
Barranca Depth Profile dating
-10.758 -77.765 53 BAR-DP1 40-50 Sand 36.2984 0.1661 3,94E-12 1.54 12.0 n.d n.d. n.d. n.d. n.d. n.d.
BAR-DP2 70-80 33.0576 0.1673 2,43E-12 3.04 82.1BAR-DP3 100-110 25.7899 0.1731 1,34E-12 2.52 60.1BAR-DP4 160-170 35.9971 0.1705 8,31E-13 3.02 26.2BAR-DP5 250-260 50.2153 0.1747 6,08E-13 5.27 14.1BAR-DP6 310-320 33.2556 0.1744 2,96E-13 4.41 10.3
Isochron burial dating
-10.758 -77.765 53 BAR-IS1-2 250-260 Quartz bearing clasts
37.67 0.1736 4,37E-13 3.07 13.4 3.55 4,35E-13 16.0 91.6 6.84 1.12
BAR-IS1-5 25.9268 0.1718 2,36E-13 4.15 10.3 3.62 2,98E-13 6.92 65.2 6.30 0.71BAR-IS1-6 32.4907 0.1741 3,13E-13 4.44 11.1 4.05 2,85E-13 6.58 62.6 5.63 1.68BAR-IS1-3 37.025 0.2017 3,11E-13 3.38 11.2 2.36 3,32E-13 11.3 67.3 5.99 0.51BAR-IS1-4 41.1355 0.2023 3,48E-13 5.25 11.3 3.04 3,83E-13 23.3 79.9 7.03 0.45BAR-IS1-7 26.5406 0.2033 2,42E-13 3.91 12.3 2.29 3,31E-13 5.38 63.9 5.21 0.35
Isochron burial dating
-10.759 -77.764 BAR-IS2-1 < 20 m Quartz bearing clasts
40.3743 0.1737 9,12E-14 8.10 2.55 3.20 3,06E-13 11.5 54.1 21.2 3.02
BAR-IS2-4 44.6497 0.1715 1,14E-13 5.99 2.85 3.94 5,91E-13 54.7 116,00 40.8 22.47BAR-IS2-6 37.5191 0.1742 1,02E-13 5.63 3.08 2.96 1,86E-13 23.5 32.9 10.6 2.58
Isochron burial dating
-10.76 -77.763 BAR-IS3-2 < 25 m Quartz bearing clasts
49.9169 0.1726 9,04E-14 5.68 2.03 4.09 2,74E-13 19.7 50.1 24.7 5.09
BAR-IS3-4 45.1403 0.1737 1,04E-13 10.11 2.61 3.42 3,11E-13 27.3 52.6 20.2 5.89BAR-IS3-7 40.7863 0.1729 7,10E-14 8.56 1,94 3.62 2,79E-13 25.5 55.3 28.4 7.68
Table 2 Input parameters for the Monte Carlo simulator in Matlab (Hidy et al., 2010).
Barranca PativilcaParameter Value Parameter Value
Latitude (degree) -10,758 Latitude (degree) -10,704Longitutde (degree) -77,765 Longitutde (degree) -77,776
Altitude (m) 53 Altitude (m) 96Strike (degree) 0 Strike (degree) 0
Dip (degree) 0 Dip (degree) 0Shielding correction factor 1 Shielding correction factor 1
Cover correction factor 1 Cover correction factor 1Uncertainty of 10Be Half-life (%) 1 Uncertainty of 10Be Half-life (%) 1
Local spallogenic production rate (at g-1 a-1) 2,50 Local spallogenic production rate (at g-1 a-1) 2,50Error in local spallogenic production rate (at g-1 a-1) ± 0.5 Error in local spallogenic production rate (at g-1 a-1) ± 0.5
Depth of muon fit (m) 6 Depth of muon fit (m) 6Error in total production rate (%) 5 Error in total production rate (%) 5
Density (g cm-3) 1.6-2.1 Density (g cm-3) 1.6-2.1X2 value 20 X2 value 3
Numbers of profiles 100,000 Numbers of profiles 100,000Age (a) 700,000-2,200,000 Age (a) 30,000-500,000
Erosion rate (cm ka-1) 0.04-0.12 Erosion rate (cm ka-1) 0.2-4Total erosion threshold (cm) 75-400 Total erosion threshold (cm) 75-400
Inheritance (at g-1) 0-70,000 Inheritance (at g-1) 0-58,000Attenuation length (g cm-2) 160 r�� Attenuation length (g cm-2) 160 r��
Table BarrancaA Results of the Monte Carlo simulations with Matlab
Age (ka) Inheritance (atom/g) Erosion rate (cm/ka)Mean 1254.3 8300 0.07
Median 1239.4 6400 0.07Mode 1197.1 300 0.07
Minimum X2 1261.5 1200 0.06Maximum 1503.0 5,8000 0.09Minimum 1043.2 0 0.05
Table PativilcaB Results of the Monte Carlo simulations with Matlab
Age (ka) Inheritance (atom/g) Erosion rata (cm/ka)Mean 200.6 3,5500 0.52
Median 200.3 3,5500 0.51Mode 204.1 3,5100 0.48
Minimum X2 217.6 3,7800 0.35Maximum 314.3 4,3800 0.89Minimum 125 2,7100 0.25
Sample namePaleo/modern erosion
ratesAge of the deposits
Latitude (DD.DD) WGS84
Longitude (DD.DD) WGS84
Altitude of the top of
the terrace (m.a.s.l)
Sample depth (cm)
Quartz dissolved (g)
9Be spike (mg)
Measured 10Be/9Be ratio
(10^-12)
AMS error (%)
10Be concentration
(at/g)
Concentration at the time of
deposition (at/g)
Denudation rates (mm/ka)
PAT-ME Modern erosion rates Modern Pativilca river 10.717°S 77.767°W 71 Surface 50.11 0.1991 0.24 4.0 6,4052 +/- 2695 6,4052 +/- 2695 260 +/- 23
PAT-DP6 Paleo erosion rates Terrace T1 (Pativilca) : 200 ka 10.704°S 77.775°W 95 465 cm 49.88 0.1743 0.24 4.6 5,6219 +/- 2634 5,5203 +/- 2539 302 +/-28
PAT-PE Paleo erosion rates Terrace T3 (Pativilca) : 40 ka 10.708°S 77.772°W 75 400 cm 49.97 0.1953 0.22 5.5 5,6468 +/- 3144 5,6004 +/- 3080 297 +/- 29
BAR-DP6 Paleo erosion rates Terrace (Barranca) : 1.2 Ma 10.758°S 77.765°W 52 315 cm 33.25 0.1744 0.30 4.4 10,2942 +/- 4574 11,6020 +/- 2464 143 +/- 12
BAR-PE2 Paleo erosion rates Terrace (Barranca) : 100 ka 10.759°S 77-764°W 33 400 cm 41.24 0.1984 0.14 6.3 4,4416 +/- 2823 4,2570 +/- 2682 392 +/- 40
top related